Re: [Bitcoin-development] Extension for BIP-0070 to support recurring payments

[Mike Hearn <mike@plan99.net>]

[-- Attachment #1: Type: text/plain, Size: 3979 bytes --]

I think the right approach for this is to actually implement it and
*then* propose
a BIP. There are so many possible features we could add to the payment
protocol, any other approach would rapidly turn into lots of people
deciding to do the "fun bits" and often leaving others doing the hard work
with difficult or unworkable specs.

For instance, if you try to implement this, you would rapidly discover that
it probably makes more sense to do this as an additional set of fields in
PaymentDetails rather than a new message type entirely. A new top level
message type would in turn require new MIME types, URI extensions and so
on. That doesn't make any sense.

Once you decide to extend PaymentDetails, the next discovery would be that
it probably makes sense to try and solve the problem of address re-use for
recurring payments first, before speccing out time intervals and so on.
That's a separate BIP.

I'm all for adding recurring payments as a feature, that's what the
protocol is there for. But I'd like to see future protocol extension
requests come after at least one working implementation has been made .....


On Tue, Jan 28, 2014 at 3:36 AM, Stephane Brossier
<stephane@kill-bill.org>wrote:

> Hi,
>
> [I sent this email 2 days ago prior my registration to the mailing list;
> please forgive me if this is a duplicate]
>
> I would like to propose an extension to the Payment Protocol (bip-0070) to
> address the case of recurring payments in Bitcoin -- new bip or
> modification of bip-0070.
>
> There has been a lot of growth in the last few years in the 'subscription
> economy' with many new companies embracing that model -- online video,
> gaming, groceries, newspapers,... In parallel, Bitcoin is growing into a
> mainstream currency (hence bip-0070), and so the next logical step would be
> to define a protocol to address that need.
>
> We have been working in the past few years on an open-source billing
> platform (http://kill-bill.org/), and recently came with a prototype to
> do recurring billing in Bitcoin (see
> http://thekillbillstory.wordpress.com/2014/01/20/bitcoin-plugin/ and
> http://thekillbillstory.wordpress.com/2014/01/11/coinbase-integration-experiment/
> ).
>
>
> The work flow would look similar to the one from bip-0070. There would
> need to be some additions; the flow could be summarized as follow:
>
> 0. Click: 'Subscribe Now'
> 1. Wallet would get  a RecurringPaymentRequestAuth which describes the
> nature of the future recurring payments
> 2. The Customer would get prompted from the wallet to authorize it.
> 3. The wallet would then poll the Merchant server (startup time, and/or
> well defined frequency) and potentially merchant would start issuing a
> PaymentRequest); the role of the wallet is to ensure that PaymentRequest is
> within the bounds of what was accepted by the customer-- amount,
> frequency,.. If it is, then it would make the Payment the same way it works
> for bip-0070
>
> Is that something that the community would be interested in? We could
> provide more details about the protocol we have in mind (messages and
> flow), and also provide an implementation with bitcoinj as a wallet and
> Kill Bill as a merchant server.
>
> Le me know what you think.
>
> Stéphane
>
>
> ------------------------------------------------------------------------------
> WatchGuard Dimension instantly turns raw network data into actionable
> security intelligence. It gives you real-time visual feedback on key
> security issues and trends.  Skip the complicated setup - simply import
> a virtual appliance and go from zero to informed in seconds.
>
> http://pubads.g.doubleclick.net/gampad/clk?id=123612991&iu=/4140/ostg.clktrk
> _______________________________________________
> Bitcoin-development mailing list
> Bitcoin-development@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
>
>

[-- Attachment #2: Type: text/html, Size: 5147 bytes --]